Constraining cosmological parameters from measurements of the Integrated Sachs-Wolfe effect requires developing robust and accurate methods for computing statistical errors in the cross-correlation between maps. This paper presents a detailed comparison of such error estimation applied to the case of cross-correlation of Cosmic Microwave Background (CMB) and large-scale structure data. We compare theoretical models for error estimation with montecarlo simulations where both the galaxy and the CMB maps vary around a fiducial auto-correlation and cross-correlation model which agrees well with the current concordance LCDM cosmology. Our analysis compares estimators both in harmonic and configuration (or real) space, quantifies the accuracy of the error analysis and discuss the impact of partial sky survey area and the choice of input fiducial model on dark-energy constraints. We show that purely analytic approaches yield accurate errors even in surveys that cover only 10% of the sky and that parameter constraints strongly depend on the fiducial model employed. Alternatively, we discuss the advantages and limitations of error estimators that can be directly applied to data. In particular, we show that errors and covariances from the Jack-Knife method agree well with the theoretical approaches and simulations. We also introduce a novel method in real space that is computationally efficient and can be applied to real data and realistic survey geometries. Finally, we present a number of new findings and prescriptions that can be useful for analysis of real data and forecasts, and present a critical summary of the analyses done to date.